Recent interviews:

Antonya Nelson: Angles of Prose (April 11, 2002)
Antonya Nelson, the author of Female Trouble, talks about her unsentimental take on the untidy worlds her characters inhabit.

Philip Ball: The Science of the Palette (April 4, 2002)
Philip Ball, the author of Bright Earth: Art and the Invention of Color, talks about the intersection of art, science, and creativity.

Jonathan Rauch: The World on a Screen (March 29, 2002)
The author of "Seeing Around Corners" talks about what the study of artificial societies has to tell us about the real world.

Jonathan Coe: Fast Times at King William's High (March 27, 2002)
A talk with the author of The Rotters' Club, a darkly humorous story of coming-of-age in 1970s England.

Theo Padnos: Teaching Behind Bars (March 15, 2002)
A conversation with Theo Padnos, who got to know teenage criminals from a unique perspective—as their teacher in jail.

Samantha Power: Never Again Again (March 14, 2002)
Samantha Power, the author of "A Problem From Hell," explores why America did all but nothing to stop the genocides of the twentieth century.

More interviews in Atlantic Unbound.

More on books in Atlantic Unbound and The Atlantic Monthly.

Atlantic Unbound | April 26, 2002
History in a Cell

Steve Olson, the author of Mapping Human History, retells the story of humanity—including the creation of different "races"—through the information encoded in our DNA


Mapping Human History: Discovering the Past Through Our Genes
by Steve Olson
Houghton Mifflin
278 pages, $25.00

he problem of the twentieth century is the problem of the color line," W.E.B. DuBois wrote in the pages of The Atlantic Monthly a hundred years ago. He turned out to be right; But will his prophecy hold true for the twenty-first century as well? Steve Olson, the author of the new book Mapping Human History: Discovering the Past Through Our Genes, thinks not. In it he writes, "Genetics research is now about to end our long misadventure with the idea of race. We now know that groups overlap genetically to such a degree that humanity cannout be divided into clear categories."

His book is a story of the last 150,000 years of human history, told through the information emerging from our DNA. In the past several decades, as DNA testing has become increasingly sophisticated, scientists have started using the historical information encoded in our DNA to learn how modern humans developed in eastern Africa more than a 100,000 years ago, how they migrated from there into the rest of the world (displacing the Neandertals and slowly splitting into groups with different appearances and customs), and how these groups then continued to mix and change as agricultural practices spread and civilizations conquered or were conquered. As Olson points out, the differences between groups are profoundly superficial, the result of just a few genes having been changed over time by different environments or cultural preferences. (For example, northern Europeans developed light hair and skin so that they could better soak in the weak sunlight in their part of the world.) According to the history inscribed in our DNA, there has been both too little time and too much mixing among peoples for groups to have diversified in deeper, more meaningful ways. As Olson writes, "The genetic variants affecting skin color and facial features probably involve a few hundred of the billions of nucleotides in a person's DNA—an insignificant amount. Yet societies have built elaborate systems of privelege and control around these minuscule genetic differences."

Steve Olson is the author of "The Genetic Archaeology of Race" (April 2000 Atlantic), much of which is included in this book. That article followed the travails of the Human Genome Diversity Project—a scientific enterprise dedicated to collecting DNA from various groups in order to study their differences. The scientists involved in the project have been caught in the middle of an essential dilemma—one that is haunting and slowing the study of our genetic history: "The only way to understand how similar we are is to learn how we differ. Yet studies of human differences can seem to play into the hands of those who would accentuate those differences." Whether we like it or not, the study of genetic variation will proceed at an ever more rapid pace. Olson's book helps us to understand the nature of that research—and to consider its implications.

Steve Olson spoke to me from his home outside Washington, D.C., on April 17.

—Katie Bacon

Steve Olsen
Steve Olson   

Have there been some common understandings of human history that have been overturned in the past few years by the study of genetic variation? If so, what have been some of the most striking discoveries?

I don't know about overturned, but in a number of cases the study of genetic variation has resolved questions that were up in the air, where human history could have occurred in several possible ways and the genetic evidence has shown that one of those ways is far more likely than the others. The classic case of that has been the out-of-Africa versus multiregional theories of modern human origins. The genetic evidence strongly argues for the out-of-Africa scenario, which would not have been the case if we were relying just on the fossil evidence.

In Mapping Human History you write that "Our basic body plan was set more than 100,000 years ago. Since then, we have been in a period of evolutionary stasis." Why wouldn't humans keep on evolving, either through genetic drift or natural selection?

I think humans have continued to evolve, but in ways that are less important than what happened in the period between a hundred thousand and a hundred and fifty thousand years ago, when modern humans appeared. When you look at human evolution, you tend to see a pattern of rapid stages of evolution followed by long periods of stasis, as you see in many other animal species. Now, this is a controversial point that's still being argued in evolutionary biology—the extent to which evolution is dominated by this kind of discontinuous change as opposed to more gradual change. But certainly the discontinuous pattern seems to fit what we see in humans. By one hundred thousand years ago, quite a different kind of human had evolved in eastern Africa than had existed in the past. According to the genetic evidence, this new kind of human then underwent what evolutionary biologists call a "range expansion"—fanning out from its place of origin all over the world. Of course, as groups have moved into places with different environments, there have been examples of less dramatic evolution. The changing of skin color is certainly the most noticeable example, but there have been other changes; for example, different groups have developed resistance to different diseases that were endemic in certain areas. Still, those kinds of changes haven't altered the basic design of modern humans that was established by a hundred thousand years ago.

In another fifty thousand years, do you think there could be something that would alter our basic design again?

Given the way human societies are now organized, I think it's very unlikely that you would ever see a substantial and dramatic evolution of another human group. Human groups mix too much, and the forces of selection that were operating a hundred and fifty thousand years ago no longer exist. I don't think humans are evolving in a major way today, and I don't foresee that they will evolve in the future, barring some catastrophe or radical change in human societies. At this point, the most likely way in which evolution is going to occur is through the conscious intervention of human beings in our DNA, and that could happen within a few generations.

You point out that many human groups are cultural groups rather than biological ones, and that many of the bitterest feuds are between groups that are not really that biologically distinct—Palestinians and Israelis, Serbs and Albanians, Hutu and Tutsi. Do you think that a greater understanding of our biological heritage would help defuse these cultural feuds?

I don't think it will necessarily defuse these kinds of feuds, because these feuds have cultural origins. People organize human groups for social and cultural reasons and then go looking for biological ways to justify the distinctions that they've made. If they can find biological reasons, they'll use them, but if the biological reasons don't exist, it doesn't really change the fact that the cultural distinctions have already been made. Then again, the way humans group themselves can have greatly exaggerated significance, because people have believed for so long that groups differ biologically. We now know that the biological differences between groups are insignificant, so maybe over time the social significance of these groupings also will decline.

How would you respond to someone who says, Well, I can see that certain groups are distinct from others, and if they look different, there must be deeper differences as well?

Certainly human groups have physical differences. But the physical differences we see tend to be relatively minor—they're facial features and skin colors. Now, if you look at the history of human beings as they expanded into the world, there are a lot of good reasons why people would develop these different physical appearances. With skin color, living in areas where people are exposed to different amounts of sunlight would be expected to change skin color for straightforward biological reasons. For facial features, different mechanisms are involved. What probably happened is that cultures developed preferences for certain physical appearances and mates were chosen on the basis of those appearances, so people with particular physical appearances tended to have more children than people with other appearances. As a result, the DNA that caused those appearances became more common within a group over time.

One could try to extend this argument for group diversification to mental differences. But you run up against two problems, which I pointed out in my Atlantic article last year. First of all, no one has ever identified a mechanism that could sort mental differences among a population as genetically connected as humans are. Second, skin color and facial features are under the control of just a few genes, whereas complex mental traits are under the control of thousands of genes. And skin color and facial features are not much affected by experience whereas all mental traits are profoundly affected by experience, from the moment of conception on. In a homogenous population like the human population, it will never be possible to ascribe the behaviors of groups to particular genetic variants, because the effects of the environment will always be much greater than the genetic differences among groups.

What sort of response do you expect to get on this book? Do you anticipate a contingent of people who are not going to be pleased with the case you're making here?

I think some people might object to particular things I say. But the interesting thing about this genetics research is how it coheres with bodies of evidence from many other disciplines. Our genetic history is not just a string of disconnected facts. It provides an overall narrative of the past one hundred and fifty thousand years, and that narrative coincides with information from other fields, including archaeology and linguistics, and from historical accounts as well. So while someone could question the validity of the evidence on any given point, I found in writing the book that the story became more and more compelling as the pieces of the overall picture fell into place.

Why do you think that the idea that there are real and fundamental racial differences between groups has been so persistent?

It's persistent in particular cultures, but not necessarily in all cultures, which suggests that it's largely a phenomenon driven by social forces. For instance, if you look at the ancient Greeks and Romans, the important distinction for them was between civilized people and barbaric people. It wasn't necessarily the color of a person's skin. The Greeks and Romans noticed skin color and speculated about it. But a person of any skin color could be accepted into Hellenistic or Roman society so long as that person spoke the language and accepted the cultural norms of the group. A person looking exactly like a Greek or Roman who did not speak the language and was not part of the culture was outside of the group. The persistence of racial thinking has been particularly strong in the New World in general, and in the United States in particular, because of the unusual history of this part of the world. Groups of people with especially different appearances from very different regions were put together and subjected to powerful social forces that led to rigid social hierarchies. So people had many reasons for dividing the occupants of the New World into groups; they had biological ways of justifying those reasons; and they have relied extensively on those reasons ever since.

Did writing this book change how you think of yourself, in terms of whether or how you identify with any particular group?

It did, because when I began writing this book, I, like many other people, tended to view human groups as fairly distinct and as the product of historically separate trajectories. I realized in the course of doing research for the book that that is an illusion. Humans as individuals and as the members of groups are much more closely related to each other than you would think. I now view the biological distinctions among groups as essentially meaningless. Culturally, of course, I'm affiliated with some groups much more closely than with others. But I see myself as no more closely related biologically to one human group than to any other. The distinctions we make are in our heads, not in our genes.

Maybe you haven't thought in this direction, but if policymakers in the U.S. were to read and act on your book, what changes would you hope would spring from it? What are the practical applications of the story you're telling?

Well, most of the policy implications have to do with biomedical applications in this work, and the biomedical applications are what's going to drive our research into human genetic differences. People are going to be gathering huge amounts of genetic information in the next few years about the differences among individuals and among human groups. I believe that a consensus exists that this biomedical research should go forward. At the same time, we could use that information to learn a great deal about the history of human groups and of individuals. What I would like to see is that these two lines of research—learning about history and learning about medical genetics—not be artificially separated but combined. That's how the biomedical information will be most useful. And it will provide information about human history that will clarify many misperceptions that people have about the relationships among human groups.

In the book you talk a fair amount about medical-consent issues, in terms of gathering data on groups. It does seem like those issues are going to be especially complicated. As you point out, it's not always easy to identify the benefits for a group of letting themselves be studied, because if you turn up a marker for diabetes, say, it could be used to deny that group insurance or hurt them in other ways. How would you make the counter-argument that groups really could benefit from being studied?

Groups will need to think carefully about how research into their genetic differences is conducted and how it's disseminated. In some cases, groups may decide that the potential risks outweigh the potential benefits. But such benefits certainly exist. For example, one may be able to look at the specific genetic variants that a group has and eventually tailor treatments to individuals and groups, though there are limits to this approach. Plus, there is the broader benefit of greater understanding of human disease in general.

I can imagine all sorts of implications once people start being able to trace their genetic heritage through their DNA. Some people might find that they're related to groups they never imagined they would be. Others might find that genetically, they don't have the heritage that they thought they did. Can we expect any sort of societal changes as more and more people learn what their true heritage is?

I think the most interesting thing that's going to happen is that there will be a much greater appreciation for the diversity of ancestors who have contributed to the DNA of any one person. This won't necessarily happen right away, because people are going to look first at the parts of their DNA that are technically easiest to analyze—namely, mitochondrial DNA and the Y chromosome. But those are very small parts of our DNA, represening just a tiny fraction of our ancestors. As we become able to look at the rest of our DNA, people are going to realize how much of a hodgepodge, how much of a genetic quilt each person actually is. That's the lesson I think will eventually emerge from studies of genetic genealogy.

So do you think in ten years I'll be able to walk into my doctor's office and say, "Okay, test my chromosomes and see what the distribution of my ancestors is."? Is that realistic?

Tests like that are being developed right now. What remains to be seen is whether anyone will find the kind of information that's generated interesting or useful. Information on your genetic background is not going to provide the kinds of certainties that a family tree in a family Bible does—in fact, it could destroy some of those certainties. People are going to find out that different percentages of their DNA come from different parts of the world. Will people be willing to pay several hundred dollars to be told something that we already know today?

The Human Genome Project, which sequenced one cell line, is very different from the Human Genome Diversity Project, which proposes to collect DNA from groups around the world. But has the Human Genome Project been confronted by issues similar to those that have dogged the Human Genome Diversity Project? If so, how has the Human Genome Project dealt with them?

I think of the Human Genome Project as a human genome project rather than the human genome project. Basically, in putting together the human genome, the organizers decided to ignore human groups—and they had good reasons for doing that. The only way to get started on the genome was to provide a base sequence and then, after the base sequence was done, to start looking at how individual sequences differ from the base sequence. Well, the base sequence has now been put together, so the people who were involved in the human genome have gone on to the next step of looking at how individuals and groups differ from each other. And that's inevitably going to be a much more controversial undertaking.

What is happening with the Human Genome Diversity Project? Has there been any change since you finished your book?

Actually, the project continues to make progress. There was a letter in the issue of Science magazine that I received yesterday notifying the genetics community that a thousand cell lines from different parts of the world are now available for researchers through an organization in Paris. The consent issues have all been taken care of for those cell lines, and a tremendous amount of research can be done just with that material. The large amounts of funding that the project might have hoped for ten years ago haven't materialized, but it continues to move along.

It sounds as though rather than moving along as a centralized project, people from all sorts of different labs are moving forward with this research individually.

A tremendous amount of work is underway, and the questions being asked are very exciting. The one great unresolved issue is the extent to which the biomedical work that's being done can be connected with people who are interested in history and anthropology. There is great potential if those areas of research were to be combined or at least coordinated somehow, but so far not much of that has been done.

Why hasn't that been done yet?

It hasn't been done so far because it's been easier not to. But it's going to become necessary. It's turning out—and people might not have anticipated this ten years ago—that the only way to thoroughly understand the genetic differences that contribute to the medical problems characteristic of a group is to understand the history of that group. As researchers continue to realize that the historical and the biomedical aspects of genetic research cannot be separated, they're going to have to figure out ways to combine these two areas of research.

Do you see them moving in that direction at all?

I do think people are moving in that direction. It's slow, but their hands will soon be forced by the pace of advances in genetics. So much information is being generated so quickly on the biomedical side that inevitably the data are going to force some decisions about merging the two fields, and as the data continue to flood in, those decisions will need to be made quickly.

Your article in the May Atlantic, "The Royal We," follows along in the same vein as your book, in that it argues that the mathematical study of genealogy is showing that we're all more closely related than we might have thought. How does the work of Joseph Chang or Mark Humphrys, the researchers in your article who are making that case, clash with conventional wisdom?

I think genealogists have usually assumed that one person's ancestors are somewhat distinct from an unrelated person's ancestors, and that's certainly true if you look at the recent past. But as you start looking at the longer-term past, you find that the ancestors of different individuals start to overlap, and that they overlap much more extensively than we'd previously thought. That's where Chang's and Humphrys's ideas clash with what's generally thought to be true about genealogy.

So what this is showing is that mathematically, there has to be more mixing than people had thought.

Oh, absolutely that's true in genealogical terms. There's no doubt that there has been a tremendous amount of mixing. After all, for a European all it takes is one ancestor who was African or Asian—of the thousand people who were your ancestors ten generations ago, or the million people who were your ancestors 20 generations ago—to connect you genealogically to an entirely different part of the world.

How does the work of Humphrys or Chang conflict with Bryan Sykes's argument in his recent book The Seven Daughters of Eve that all Europeans are descended from seven "clan mothers"?

That's where the genetic and the genealogical intersect. You can look at Europeans and say that 95 percent of them have mitochondrial DNA from just seven women who lived in the past, and so Bryan Sykes, a geneticist in England, has called these women the "seven daughters of Eve." You could therefore conclude that individual Europeans are descended from these seven women in some meaningful way—in other words, that these women had some important role in the ancestry of each European. But the fact of the matter is that all Europeans today are descended from almost all of the people who lived at the various times when these seven women lived. It's a statistical artifact that a given person would happen to have mitochondrial DNA from one of those seven women. It says little or nothing about ancestry in any realistic sense because in fact all Europeans are descended from all seven of those women. For that matter, so is everyone else in the world.

You point out that intermarriage between different groups is only becoming more common. Ultimately, what do you think this will mean for people's conceptions of ethnicity and race?

In the final chapter of the book I talk about Hawaii, where intermarriage is quite common and has been for some time. In Hawaii you still have ethnic and racial tensions between groups, and those tensions aren't going to go away any time soon. But my experience has been that those tensions have a different feel in Hawaii. There's more recognition that these groups are culturally and socially defined, so there's more flexibility in ethnic affiliation. People have more freedom to be members of particular groups regardless of their actual genetic background. Now, it's not the case in Hawaii that you have complete freedom to choose your ethnicity, but that's the direction in which things are headed. And when you look at the genetics of human groups, that idea makes perfect sense. People have been so tightly connected throughout their history that you can't divide people up genetically into these rigid groups. It would make much more sense if ethnicity were a sort of voluntary affiliation, like religion or a profession. If that were the case people might quit looking for some sort of mysterious biological essence inherent in particular groups and instead acknowledge that all human groups are biologically linked.

What do you think? Join the conversation in Post & Riposte.

More interviews in Atlantic Unbound.

More on books in Atlantic Unbound and The Atlantic Monthly.

Katie Bacon is an editor for The Atlantic Online. Her most recent interview was with Jonathan Rauch.

Copyright © 2002 by The Atlantic Monthly Group. All rights reserved.